New oblique impact methods for evaluating head injury mitigation effects of helmets are emerging, which mandate measuring both translational and rotational kinematics of the headform. These methods need headforms with biofidelic mass, moments of inertia (MoIs) and coefficient of friction (CoF). To fulfil this need, the working group 11 of the European standardization head protection committee (CEN/TC158) has been working on the development of a new headform with realistic MoIs and CoF, based on recent biomechanics research on the human head. In this study, we used a version of this headform (Cellbond) to test a motorcycle helmet under oblique impacts at 8m/s at five different locations. We also used the Hybrid III headform, which is commonly used in helmet oblique impacts. We tested whether there is a difference between the predictions of the headforms in terms of injury metrics based on head kinematics, including peak translational and rotational acceleration, peak rotational velocity and BrIC (Brain Injury Criterion). We also used Imperial College finite element model of human head to predict strain and strain rate across the brain and tested whether there is a difference between the headforms in terms of predicted strain and strain rate. We found that the Cellbond headform produced similar or higher peak translational accelerations depending on the impact location (-3.2% in front-side impact to 24.3% in rear impact). The Cellbond headform however produced significantly lower peak rotational acceleration (-41.8% in rear impact to -62.7% in side impact), peak rotational velocity (-29.5% in side impact to -47.6% in rear impact) and BrIC (-29% in rear-side impact to -45.3% in rear impact). The 90th percentile value of the maximum brain strain and strain rate were also significantly lower using this headform. Our results suggest that MoIs and CoF have significant effects on headform rotational kinematics, and consequently brain deformation, during helmeted oblique impacts. Future helmet standards and rating methods should use headforms with realistic MoIs and CoF (e.g., the Cellbond headform) to ensure more accurate representation of the head in lab impact tests.